1877.] 



Supersaturated Saline Solutions. 



527 



on the surface to be tried, and around this were arranged a number of 

 single drops. My observation agrees generally with that of Prof, 

 Grenfell, that the large drop is more ready to solidify than the smaller 

 ones ; and the reason that I should'give for this is, that the larger surface 

 is more likely to catch nuclei from the air than the smaller ones. In the 

 early stage of this inquiry it was thought that the sudden shock pro- 

 duced by the inrush of air when the bladder was penetrated produced 

 crystallization ; but Prof. Turner, of University College, London, showed 

 that if the neck of the flask containing the solution were drawn out to 

 a capillary opening the solution could be retained in the flask without 

 becoming solid. Lowel also showed that solutions could be kept longest 

 in ordinary flasks in proportion as their necks were narrow ; and another 

 early observer noticed that while a flask with its neck upright led to 

 crystallization, it was less likely to do so if the neck were placed at an 

 angle, and still less so if placed horizontally. This evidently points to 

 the increased difficulty of a nucleus getting into the flask under such cir- 

 cumstances. 



I tried a large number of experiments on supersaturated solutions in 

 Prof. G-renf ell's manner, not only with those of sodic sulphate, but also 

 of sodic acetate, magnesic sulphate, zinc sulphate, potash alum, ammonia 

 alum, sodic carbonate, and some others. On uncovering the flasks for the 

 purpose of filling the dropping-tube, the solutions in the flasks often 

 crystallized suddenly * ; and it has not unfrequently happened that the 

 drop of sodic sulphate and of some other solution at the end of the 

 dropping-tube suddenly became solid before it fell, and thus acted as a 

 nucleus to the contents of the tube. 



I repeated Prof. G-renfell's experiments with oily surfaces. A drop of 

 oil (olive or niger) rubbed with the finger over a glass plate made the 

 plate powerfully nuclear : drops of sodic sulphate became immediately 

 solid, the large drop swelling up into a well-shaped conical figure with a 

 rounded top. 



Experiments with magnesic sulphate (2 to 1 solution) showed this salt 

 in drops to be even more sensitive than sodium sulphate. Drops of zinc 

 sulphate (2 to 1) and of sodium carbonate (3 to 1) soon became solid on 

 glass, paper, and other surfaces, whether in the open air of my garden 

 or in different rooms of my house. 



Potash alum solution (2 to 1) in drops crystallized rapidly into an 

 opaque white mass, or into transparent crystals in concentric circles 



* In reboiling, the flask must be kept in motion over the flame so as to melt the salt 

 gradually and not allow one part of the flask to become much hotter than another. 

 As the solution approaches boiling, the flask, being chemically clean, is liable to violent 

 bumpings. These may be prevented by keeping in each solution a fragment or two of 

 charcoa (that from cocoanut-shell being most efficacious), as pointed out in my paper 

 in the ' Proceedings ' for 1869, p. 251. That the action of the charcoal is not impaired 

 by use has been shown in these repeated reboilings as well as in the cases given by me 

 in the Phil. Mag. for August 1875. 



